Interest Groups and Policy Adoptions in the Late 1800s

The power and influence of politically-oriented associations, often aided by federated membership structures paralleling the organization of American government, has been well-documented across a broad range of issue areas.Footnote ² On agriculture (the Grange, Farmers’ Alliance, Colored Farmers’ Alliance), veterans’ benefits (the Grand Army of the Republic) labor rights (the Knights of Labor, the American Federation of Labor), and women’s rights (the General Federation of Women’s Clubs, National American Woman Suffrage Association), associations at the forefront of creating, advocating, and helping to pass legislation.

In many cases, these groups achieved substantial success lobbying Congress. Theda Skocpol and co-authors provide evidence of the General Federation of Women’s Clubs’ successes in achieving the passage of mothers’ pensions in the early twentieth century, and Scott Ainsworth demonstrates the important role played by the Grand Army of the Republic in increasing Union war veterans’ pensions.Footnote ³ Yet, much of the legislative work during the late nineteenth century was conducted at the state level. Interest groups likely played a direct role in aiding the passage of laws in state legislatures as well, but less is known about their mobilization and successes in these institutions.

Historical evidence favors an active role for interest groups for several reasons. First, late nineteenth-century state legislatures were not institutionalized and lacked professional members and staffs. There were no professional organizations of legislators, a greater number of bodies met on a part-time basis, and membership turnover was high.Footnote ⁴ Second, while partisanship clearly affected how state legislatures behaved during this era,Footnote ⁵ nineteenth-century state governments were viewed as “spare, with little administrative muscle.”Footnote ⁶ If state policy was going to shift, the impulse was unlikely to come solely from legislative parties, as the legislatures were outdated bodies that “could not lead the citizenry toward a more progressive future.”Footnote ⁷ Governors took up some of this slack,Footnote ⁸ but clearly, the high volume of legislation produced by state governments of this era did not come from governors alone.Footnote ⁹ In this environment, interest groups could enter the fray and provide the expertise and the push to turn proposed bills into laws.

Furthermore, national associations interested in particular legislative issues could then serve as a conduit for policy adoptions across the states. These associations attracted significant numbers of members and developed extensive communications networks and professional staffs trained to advocate for policy change—a task that citizen legislatures were ill-equipped to perform. But the policy diffusion literature provides only a few ideas about how groups may have filled this void. In fact, despite over eight hundred articles on policy diffusion written since Frances Stokes Berry and William D. Berry’s seminal work on state lotteries, relatively little is known about policy diffusion prior to the twentieth century.Footnote ¹⁰

This said, scholarship on modern policy diffusion provides insights that may help us to better evaluate why states adopted policies during this era. Many of these studies emphasize the role of geography in policy diffusion, whether through neighboring statesFootnote ¹¹ or regional patterns of policy adoptions.Footnote ¹² Yet some scholars contest the importance of geography and find it less important than other factors such as state ideologyFootnote ¹³ and partisanship.Footnote ¹⁴ Still others emphasize the importance of interest groups in spreading policies. Models capture this interest group participation in a variety of ways, including counts of lobbyists,Footnote ¹⁵ scholar-created measures of group concerns about policy areas,Footnote ¹⁶ membership,Footnote ¹⁷ and even model legislation.Footnote ¹⁸ The types of policies also matter in this process, with state governments learning from other governments on nonmorality policies while bending to the demands of citizens on morality policies.Footnote ¹⁹

Taking into consideration the rise of federated voluntary associations, the nature of state governments, and the prospects for diffusion given the context of the era, it appears that the environment was ripe for membership associations to push reform demands through state legislatures. These early interest groups spanned the nation, and their federated structure likely allowed them to advocate for policies in state legislatures by relying on support from both national leaders and state-level membership. Unified state legislatures, or bodies where both houses were controlled by the same political party, were likely good targets, especially on issues of morality, where political parties appear more likely to bend to perceptions that citizens want change. And, efforts to influence states to adopt policies were potentially aided by geographic proximity. Even during an era where modern communication technology was still developing, neighboring legislatures learned from one another; legislators observed whether policies caused controversy in government and among the public, or garnered sympathy throughout the legislature and state. Interest groups served as a key conduit for this information, and thus, were likely key motivators of policy adoption.Footnote ²⁰

The Three Policy Areas and the WCTU

To assess this theoretical portrayal, we consider a single major voluntary association that had broad capacity to get its policy goals adopted—the Woman’s Christian Temperance Union (WCTU). Founded in 1874 as a women’s-only organization seeking to eradicate alcohol consumption and the liquor traffic, its early efforts focused solely on temperance and prohibition. Under its most notable president, Frances E. Willard, the group’s agenda broadened to a “Do Everything” approach, including concerns related to “home protection”: clean living, education, safeguards for women and children, and opposition to vice and obscenity. The WCTU used a variety of direct and indirect lobbying techniques to influence government policies on these subjects. In particular, the group was the leading organization petitioning the 55th Congress (1897–98) on moral issues.Footnote ²¹ However, the strategies it employed depended on the issue and the venue.

Here we consider three policy areas critical to Willard’s “Do Everything” approach that required lobbying government: mandating scientific temperance education, increasing the age of consent, and barring tobacco purchases by children. The WCTU was a driving force for reform, and, as shown in Table 11, between 1882 and 1901 the vast majority of states had acted on all three policy areas, consistent with WCTU demands. Furthermore, a clear majority of these reforms were passed through unified and generally Republican-controlled state legislatures.Footnote ²² If we consider the years up to and including the first twenty states to adopt a law, 74 percent of scientific temperance instruction laws (1882–87), 78 percent of age-of-consent increases (1885–89), and 73 percent of anti-tobacco laws (1883–89) were passed by unified Republican legislatures. Divided legislatures passed 9 percent of age-of-consent laws and 4 percent of laws in each of the other areas. This means unified Democratic legislatures were adopters under 25 percent of the time (22 percent, 13 percent, and 23 percent, respectively). The descriptions below highlight what scholars know about the WCTU’s efforts to achieve passage of these bills.

We begin with the policy area most connected to the WCTU’s core mission of spreading anti-alcohol sentiment: scientific temperance education. The goal of this program was to influence the education of America’s youth by legislating that students needed to be taught about the dangers of alcohol, supported not by moral claims, but by those of doctors and scientists. This “scientific” approach would then shape a new generation of Americans taught to abstain from drinking and, potentially, serve as advocates for prohibition.

Historical work has noted the connection between Mary Hunt, superintendent of the WCTU’s Department of Scientific Temperance Instruction (STI), and the rapid expansion of this policy. To succeed in this effort, “Mrs. Hunt designed a plan of action to acquire the necessary legislation. The plan stipulated that WCTU women should do all in their power to put pressure upon legislators and to have pro-temperance candidates nominated in election years.”Footnote ²³ Using the STI Department’s resources, Hunt recruited state and local members to help implement her plan, which included staging “mock school boards” to prepare women for direct lobbying, sending women to schools to evaluate scientific temperance instruction, and setting up meetings where carriages, sent and paid for by the WCTU, brought local lawmakers to listen to their demands.Footnote ²⁴ The plan worked well, with early successes in Vermont (1882), Michigan (1883), and New Hampshire (1883).Footnote ²⁵

Despite these successes, Hunt’s goal was nationwide adoption, and she believed that winning approval in several key states was the lynchpin to the WCTU’s success. Thus, in a report to the WCTU, Hunt argued that “a wise general in planning a conquest aims at strategic points.”Footnote ²⁶ In this case, her targets were New York (1884) and Pennsylvania (1885). The logic was straightforward: “If these great centers of population could be carried, other smaller States would, with less labor, fall in line.”Footnote ²⁷ Though the battle in Pennsylvania was more difficult, the strategy worked; every state passed STI legislation by 1901. In states that adopted weaker policies, the WCTU continued to lobby for further legislative changes to strengthen existing laws. However, backlash from some in the scientific community concerning the veracity of arguments made by Hunt and in Hunt-approved textbooks,Footnote ²⁸ and from some educators, ultimately limited STI’s implementation.Footnote ²⁹ Still, scientific temperance instruction was so broadly adopted that it touched millions of Americans in some way. By most accounts, Hunt’s leadership was the centerpiece of the policy adoption campaign, and her efforts can be described as those of a policy entrepreneur providing a legislative subsidy.Footnote ³⁰

Second, the WCTU sought to protect women and children, and especially young women. As such, the association demanded states increase the age of consent for sexual relationships. When the WCTU began to focus attention on the issue in the 1880s, the age of consent in most states was ten or twelve years of age.Footnote ³¹ The New York Committee for the Prevention of State Regulation of Vice was among the first to lobby for an increase in 1886 after receiving a WCTU report by member Georgia Mark.Footnote ³² Subsequently, the WCTU became the main driver of policy adoption on this issue. As Mary E. Odem wrote, “Because of its extensive political network, the WCTU was better able than any single organization to build a formidable national campaign that touched every state in the country.”Footnote ³³

Beginning in 1885, the WCTU’s Department for the Promotion of Social Purity rapidly expanded into state and local unions. Through petitions, lobbying, and by arousing public sentiment, the WCTU believed it could be the catalyst for changing the age of consent nationwide. In 1886, for example, a sample petition for contacting Congress and state legislatures to raise the age of consent to eighteen was published in President Frances Willard’s Annual Address.Footnote ³⁴ And, while, as Willard asserted, “This work must not, in the least, supersede our main effort, which is for prohibition,”Footnote ³⁵ the effort proved quite successful, with most states increasing the age of consent by 1900 and all states doing so by 1920.Footnote ³⁶ This is why political scientist James Morone wrote, “In its struggle to raise the age, the WCTU won some of its most unambiguous victories for home protection,”Footnote ³⁷ and why other scholars argue that raising the age of consent was a true success for the WCTU.Footnote ³⁸

Third, the WCTU attacked another vice: tobacco use among children. As an outgrowth of their quest for a “pure” society, the association targeted tobacco generally and cigarettes specifically, though not necessarily with the same aplomb as they attacked alcohol or protected female chastity. Nevertheless, scholars note that the WCTU and “its splinter organizations were the requisite forces behind the initial cigarette prohibition movement in both the United States and Canada.”Footnote ³⁹ The WCTU formally organized against tobacco in 1883 through its Department of Narcotics.Footnote ⁴⁰

The WCTU had its greatest success in this area lobbying for legislation banning the sale and/or distribution of tobacco to children. On this subject, Marc Linder’s voluminous work on anti-cigarette laws squarely places the WCTU in the center of the drive for anti-tobacco legislation aimed at minors, prior to the association’s efforts to prohibit cigarettes entirely.Footnote ⁴¹ So connected was the group to this legislation that New York state legislators passed a law restricting tobacco sales to minors with the understanding that the WCTU would help enforce the law.Footnote ⁴² And, the group at least tried to oblige; reports from the Department of Narcotics reveal that, after petitioning legislatures to change the laws successfully, women were encouraged to help enforce the laws by monitoring their local communities.Footnote ⁴³

Taken together, these three issue areas demonstrate how President Frances Willard’s calls to “Do Everything” motivated leaders and activists to write legislation, pressure politicians, and influence other associations and citizens to communicate with policymakers. Ultimately, the WCTU was successful at converting these policy goals into legislative successes, but exactly how and why the WCTU was able to achieve these victories merits further investigation.

All three of these issues are largely moral issues, and as Gaines Foster notes, this type of “legislation had become a matter of pressure or interest-group, rather than partisan, politics.”Footnote ⁴⁴ Thus, state WCTU membership might matter, as a larger number of active women, petitioning the legislature and contacting politicians, could have caused legislators to react. In addition, modern findings suggest that stronger citizen groups can push through policy adoptions more easily on moral issues than nonmoral ones. Yet legislation is typically easier to pass through a unified legislature. The evidence in Table 1 suggests that perhaps the WCTU targeted states with more receptive legislatures first; these often were Republican-led legislatures. Although A. Lawrence Lowell’s classic study suggested that party-line voting was not that high in state legislatures of the era,Footnote ⁴⁵ there is evidence suggesting that Republican legislators tended to be more supportive of morally-focused legislation than Democratic ones.Footnote ⁴⁶ Additionally, perhaps there was some effect from neighboring states too because Republican-led legislatures tended to be located near other Republican-led legislatures. If a proximate state had already passed legislation in an issue area, it could have helped the WCTU’s lobbying efforts by empowering the women and inspiring legislators on the subject. Given these possibilities, we must move beyond historical accounts alone and to a quantitative approach in order to more fully decipher which factors mattered most to the WCTU’s lobbying efforts.

Table 1. Year of Passage, by State and Type of Legislation

* Sources: Various publications. BOLD states are those with unified Republican legislatures. Italicized states are those with divided governments. Others are Democratic legislatures.

Data and Models

To begin, we collected data on policy adoptions from various sources. Data on the passage of scientific temperance came largely from Mary Hunt’s own work,Footnote ⁴⁷ supplemented by WCTU reports and state legislative journals for the remaining states (Arkansas, Virginia, Georgia, and Utah). Age-of-consent laws were taken, in part, from Susan B. Anthony and Ida Husted Harper’s The History of Woman Suffrage, Volume IV,Footnote ⁴⁸ as referenced in Mary E. Odem’s work,Footnote ⁴⁹ with additional guidance from David Pivar and state records.Footnote ⁵⁰ Because what counted as an “age-of-consent” law varied in the eyes of activists,Footnote ⁵¹ we paid particular attention to laws that defined as rape all sexual activity occurring prior to a particular age.Footnote ⁵² These statutes typically increased the age of consent from the English common-law standard of ten or twelve, with the exception being Oregon, where the age was fourteen at the start of the WCTU’s efforts. Laws prohibiting the sale of tobacco to minors were partially listed in a publication by Clark Bell,Footnote ⁵³ though due to its incomplete nature, we checked and obtained data on other states from legislative records.

To estimate the probability of bill passage in a given year, we construct a state-year dataset, with each observation either coded as adopting (1) or failing to adopt (0) the policy. This is consistent with modeling in the policy-diffusion literature. This data structure allows us to estimate logit models.Footnote ⁵⁴ Upon passage, a state drops out of the dataset. For these three policy areas and during the time period studied, no state eliminated a law once passed, so no state reappears again after passing a law. Territories that became states are included in the dataset only after statehood. We also model the three policy areas separately because historical evidence suggests that the WCTU’s tactics were not identical across the three areas. Thus, we expect that the effects of our predictors may vary with issue area.

In studying bill adoption, the timing of legislative meetings is important. Some state legislatures met yearly, while others met biennially. Some biennial legislatures, moreover, still met yearly through adjourned sessions. And during the time period we studied, some states altered their meeting dates and meeting frequencies. Previous diffusion models have handled these concerns in three ways: a dummy variable for no legislative session, eliminating state-years when there was no meeting, or acknowledging the problem but moving forward without any controls.Footnote ⁵⁵ Here we choose only to include the years in which a state legislature met at least once, determined by the presence of state legislative reports available in archives.Footnote ⁵⁶ This way, a state-year in the models truly indicates a real possibility of bill passage.

We consider three key predictors that allow us to examine the determinants of the WCTU’s legislative success. First, absent modern measures of lobbying registration, we measure interest-group strength using state WCTU dues. These data were obtained from the national proceedings of the WCTU from 1882 to 1901. We use this information to calculate two measures of state WCTU strength: state dues paid to the national body on a per-member basis, and a per capita measurement of state dues. The former is a raw measure of association size, measured in total dollars given to the national body; it ranges from 0 to 2,251.4, with a mean of 307.07 and a standard deviation of 425.22.Footnote ⁵⁷ The latter measures the concentration of the WCTU relative to the state’s total population, measured by dividing WCTU dues by state population and multiplying by 1,000 to create a scale ranging from 0 (no group presence) to a maximum of .94 (with a mean of .22 and a standard deviation of .19).Footnote ⁵⁸ Either of these variables could have a potentially significant effect on policy adoptions. On one hand, it is possible that the states with the largest WCTU associations had greater political power, and thus were better equipped to lobby state governments to change policy. On the other hand, it may be that per capita WCTU strength allowed groups to have greater clout in the legislature as well as facilitating resource sharing with other states’ unions.

Second, we consider whether the state legislature was under unified under Republican control (1) or not (0); this measure was calculated from Michael J. Dubin’s work on the partisanship of state governments.Footnote ⁵⁹ We believe unified control, which typically facilitates agreement and expedites legislative enactment, will be particularly important in cases where the Republican Party controls both houses of the legislature, since its members were likely more sympathetic to morality politics, in general, and the WCTU, in particular. It is important to underscore that national and state Republican parties were not the originators of these laws. However, the Republican Party was much more receptive to women’s participation on some subjects (including education and the protection of women), and the WCTU’s white, Anglo-Saxon, Protestant image correlated well with the Republican Party’s demographic base. Furthermore, even when the WCTU formally supported the Prohibition Party, the organization was quite Republican in orientation.Footnote ⁶⁰ This is evidenced by the WCTU working closely with Republicans, including Henry W. Blair of New Hampshire, the first representative to introduce a prohibition amendment in Congress. Blair of New Hampshire, the first representative to introduce a prohibition amendment in Congress.

Third, we consider the proportion of neighboring states that adopted the policy in previous years.

These proportions are calculated by the authors and range from 0 (no neighboring states adopted) to 1 (all neighboring states adopted).Footnote ⁶¹ These proportions are somewhat more difficult to measure in this era because of territories. For scientific temperance, we include territories in the measurement because the national government passed a scientific temperance law that applied to all territories in 1886. This national intervention potentially influenced the behavior of neighboring states. But because the federal government did not intervene on age-of-consent laws or anti-tobacco legislation for minors, and because territories were not states, we do not include territorial action on these subjects in neighboring state calculations.

Finally, to control for the likelihood of passage following both linear and nonlinear trends, we also include time trend and time-trend squared variables, which begin in the first year of bill passage.

Including these indicators is common practice in the diffusion literature and is intended to ensure that the effects of neighboring state adoptions, which increase over time, are modeled independently.Footnote ⁶²

Empirical Results

The first set of models is presented in Table 2.Footnote ⁶³ These models confirm what Table 1 suggested: unified Republican governments are always positive, significant predictors of policy adoption. This is consistent with the WCTU’s relationship to the party; the association benefitted from legislative bodies more receptive to its goals. The effect of Republican legislatures on policy adoption, along with the relative weakness of WCTU state organizations, also explains why adoptions in all three areas were slower to occur in the South, as shown in Table 1.Footnote ⁶⁴ As alternative model specifications shown in Appendix Table A reveal—which include a dummy variable for the South (states that seceded during the Civil War were coded as 1)—these delays were not the result of a state simply being southern and the social and cultural differences particular to the region.

Table 2. Diffusion of Scientific Temperance Instruction, Age of Consent, and Anti-Tobacco for Children Legislation

* p<.05, two-tailed test. Models are logistic regression. Scientific temperance models range from 1882 to 1901; age of consent, 1885 to 1901; anti-tobacco for children, 1883 to 1901.

To investigate the predictive power of Republican control, we estimated predicted probabilities with all other predictors at their mean values.Footnote ⁶⁵ In non-Republican legislatures, the probability of adopting scientific temperance legislation is .08; in a unified Republican legislature, it increases to .49. The strength of this effect is likely at least in part due to Mary Hunt’s lobbying efforts; she ushered most of these bills through state legislatures personally, and her choices were potentially made with partisanship in mind. We see similar, although more modest, changes in the other policy areas. For age-of-consent, the increase is from .08 to .29; for anti-tobacco for children, it is from .05 to .19. Substantively, then, in unified Republican legislatures, our policies had approximately one-in-two, one-in-four, and one-in-five chances of passage. In nonunified Republican legislatures, these probabilities dropped below one in ten.

Our initial models, however, provide little support for our expectations regarding interest-group strength and neighboring state adoptions. WCTU membership is only significant in the age of consent, unadjusted dues model. Neighboring state adoptions are never significant. For both scientific temperance and anti-tobacco legislation, then, unified Republican legislatures are the best predictor of successful adoption. For age of consent, unadjusted WCTU dues also played an important role in predicting the probability of passage.

It appears, then, that state legislative venue shopping was an important element of the WCTU’s plan to achieve moral reform and home protection. But we must also consider if, in the absence of predictors for state legislative control, there were systematic secondary factors that influenced the success of WCTU-supported policies.Footnote ⁶⁶ It may be, for example, that these models reveal evidence that geography and the strength of a state’s WCTU also influenced the likelihood of policy adoption. These models are shown in Table 3.

Table 3. Diffusion of Scientific Temperance Instruction, Age of Consent, and Anti-Tobacco for Children Legislation, Excluding Legislative Control

* p<.05, two-tailed test. Models are logistic regression. Scientific temperance models range from 1882 to 1901; age of consent, 1885 to 1901; anti-tobacco for children, 1883 to 1901.

There are important differences in our three cases. The adoption of scientific temperance instruction is never significantly affected by WCTU strength, though the proportion of neighboring states that adopted is significant in the unadjusted WCTU dues model. For age-of-consent and anti-tobacco legislation, both WCTU dues measurements are positive, significant predictors of passage, though neighboring state adoptions never reach statistical significance at the p<.05 level. Thus, aside from scientific temperance, which appears to be a unique case, we find support for our expectations about the role of interest groups in nineteenth-century policy diffusion. However, we find much more limited support for geographic diffusion:

To visualize these relationships, we plot marginal effects using the unadjusted WCTU dues model for scientific temperance instruction (Fig. 1), and both WCTU models for age-of-consent (Fig. 2) and anti-tobacco legislation (Fig. 3). Time trends were set to four years for scientific temperance (1885), three years for age-of-consent (1887), and seven years for anti-tobacco legislation targeted at children (1889), which were the years when each policy’s likelihood of passage was strongest.

Figure 1. Scientific temperance instruction, marginal effects plot for neighbor adoptions

*Calculated from Table #3, Scientific Temperance Model #1.

Figure 2. Age of consent, marginal effects plot for WCTU state association dues to the national organization (unadjusted and per capita)

*Calculated from Table 3, Age of Consent Models #1 and #2.

Figure 3. Anti-tobacco for children, marginal effects plot for WCTU state association dues to the national organization state association dues to the national organization (unadjusted and per capita)

*Calculated from Table 3, Anti-Tobacco Models #1 and #2.

Other variables were held to their mean values.

Figure 1 illustrates how the probability of adopting scientific temperance legislation changes as neighboring states adopt the law, with WCTU dues held at its mean value. When no neighboring states have passed a law, the predicted probability of bill passage is .11; when 60 percent of neighboring states have done so, the probability of adoption increases to .33; and if all neighboring states have passed the law, the probability increases to .56. But confidence intervals overlap, except at the two extremes of no neighbors and all neighbors having passed scientific temperance legislation. In short, there is evidence of some neighbor adoption effect in this policy area.

Figure 2 reveals that a state’s probability of passing an age-of-consent law is strongly affected by both types of WCTU dues. WCTU membership clearly affected policy adoptions in this area, as seen in the left part of the figure—note that this was also the one instance in where WCTU dues mattered when unified Republican legislatures were considered. When state payments to the national body were $100 or less per year (1,000 members), the probability of bill adoption was around .11; when these payments hit $400 a year (4,000 members), the probability jumped to .44; and in strong WCTU states that paid $700 or over (7,000 members), the probability was .84 or higher. But per capita state dues to the national body also mattered, as shown on the right side of the figure. Though the strength of its effect is not as dramatic, and there is more overlap in the confidence intervals, as one moves from a state with per capita WCTU dues of .2 (near the mean value on the adjusted scale) to .5 (between one and two standard deviations above the mean), the probability of bill adoption increased by .21. Thus, we can conclude that interest-group size, and its relative strength to some degree, influenced state legislative adoption of age-of-consent laws. The presence of a public opinion effect on this legislative issue may owe to its particularly moral nature.

Figure 3 replicates the approach taken in Figure 2, with unadjusted dues on the left side of the figure and per capita dues on the right. Once more, as WCTU dues increase, the probability of an anti-tobacco bill being adopted increases. With unadjusted dues at $100, the probability of bill adoption is approximately .3; with dues around $400, the probability is .38; and at $700 or over, the probability is .48 or higher, although confidence intervals reveal some uncertainty in the predictions. Here, however, the probability change is stronger with per capita dues. The probability of bill adoption increases by .24 as one moves from per capita dues of .2 to .5. Thus, for tobacco legislation, it was the WCTU’s relative strength in a state, and to some degree its sheer size, that affected policy adoption. This suggests that the group may have been strategic in thinking about its legislative influence, both in terms of the capacity to have members appear at statehouses and sign petitions and also their ability to claim to speak for a significant percentage of the state’s citizens.

Discussion and Conclusion

Overall, these results shed light on the conditions that made the Woman’s Christian Temperance Union’s efforts at policy reform successful when it came to scientific temperance instruction, increasing the age of consent, and banning youth from purchasing tobacco products. First and foremost, the prominent role played by unified Republican legislatures—especially when combined with qualitative evidence—suggests that the WCTU sought out venues that were likely to be receptive to its message. This is clearly underscored in Table 1, and for scientific temperance education, where Republican legislatures had a one-in-two probability of adopting the policy in any given session.

Second, empirical evidence shows that this focus on unified Republican legislatures was supported at times by geographic proximity and membership. For the former, Mary Hunt intentionally targeted states that she thought might influence other states to adopt scientific temperance legislation, and this appears to have worked to some degree.Footnote ⁶⁷ While we cannot ascertain with available evidence if she intended to target specific Republican legislatures first to ease her efforts in neighboring states, the empirical results suggest that unified Republican legislatures were more sympathetic to Hunt’s efforts and that these legislative victories made passage more likely in neighboring states. For the latter, stronger state-level membership—whether absolute or per capita—helped the WCTU’s efforts to convince state legislatures to adopt increases in the age of consent and to ban youth from purchasing tobacco products. Since petitioning was a key component of the WCTU’s lobbying strategy for both issues, having a larger pool of workers and potential signatories was important.

In summary, the present analysis reveals the key factors that allowed a large, politically-oriented federation like the WCTU to successfully spread its policies across the United States. As a case study of a single association, though, the findings here needed to be tested on a broader range of issues and associations. Here, we consider a women’s association’s efforts to pass policies related to the education and protection of children. It may be that this altered the dynamics of policy adoption. Therefore, future scholarship on nonmorality policy adoptions during this era appears to be a fruitful avenue for future research. Although it seems likely that state legislative partisanship, group strength, and geographic proximity would influence diffusion in other policy areas, too, the relative impact of these factors may vary with issue area. The importance of these factors is likely regulated by counter-mobilization against policy adoptions, which should also be taken into account. While the policies studied here were successfully adopted by almost all states over a relatively short period, and evidence suggests resistance was present but not necessarily aggressive or even mobilized (anecdotally, age-of-consent legislation appears to have faced the most backlash), future studies should place more emphasis on gauging how forces opposed to policies succeeded in stemming, or failed to stem, the tide of their adoptions.

It may also be that a women’s group lobbying for policy change—by the very virtue of its members’ disenfranchisement—was regarded differently than other associations of the time. Did men’s voluntary associations, or associations allowing both men and women to join, employ similar or different tactics? Additionally, did other women’s associations see the same levels of success when pursuing state policy adoptions? And what about business associations, which came on the scene during the late nineteenth and early twentieth centuries?Footnote ⁶⁸ What methods did they use, and was diffusion easier or harder for them? Once more, though, it seems probable that men’s and business associations would consider state legislative partisanship, the strength of their organizations in a state, and the policy environment in neighboring states when considering which states to emphasize in their quests to get desired policy goals turned into law. The approach used here to study the WCTU and these three policy areas should be applicable to a wider sample of associations and laws, and the results presented here provide future scholarship with guidance on how to study policy adoptions during this era.

In total, it appears that the Woman’s Christian Temperance Union, whether through the leadership of policy elites such as Mary Hunt or the efforts of its general membership, was able to strategically mobilize first in states with unified Republican state legislatures to win policy adoption in all three policy areas studied. This shows how the largest woman’s voluntary association of the late nineteenth century was a major political force, able to play a role in motivating state governments to adopt legislation key to its home protection agenda. In the end, the WCTU understood how to be successful in lobbying and was able to “Do Everything” because of it.